The Agent Skills Directory

[SAFE]: The skill operates entirely as a creative direction layer and explicitly prohibits writing or executing implementation code. All workflows involve high-level strategy and delegation to internal sub-agents (Muse, Palette, Forge, etc.) rather than external scripts or sensitive system commands.
[SAFE]: Security guardrails are deeply integrated into the core methodology. The skill enforces WCAG 2.2 AA accessibility standards, mandates explainability for AI-driven interfaces, and prohibits 'dark patterns' or 'prediction-driven UI' without user override.
[SAFE]: External references (such as Apple HIG, W3C/WCAG documentation, and well-known design industry sources) are used for legitimate research and standards compliance. No suspicious remote code execution or data exfiltration patterns were detected.

vision